Geng Hong
5 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The paper introduces Tree structured Injection for Payloads (TIP), a novel black-box attack framework that reliably generates stealthy injection payloads to seize control of LLM agents utilizing the Model Context Protocol (MCP), even when defenses are present.
This paper systematically analyzes the resilience of LLM-enhanced search engines against black-hat SEO attacks, finding that while they block most traditional attacks, they remain vulnerable to sophisticated LLM-generated query manipulations.
AgentGuard is an attribute-based access control framework designed to mitigate severe security risks, such as privacy leakage and system compromise, in tool-using LLM-based agents.
The authors evaluate the autonomous penetration capabilities of 19 LLMs in a realistic cyberattack scenario.
The paper introduces AgentCyberRange, an open, multi-range infrastructure for measuring autonomous cyber attack capability in realistic cyber ranges, and evaluates six frontier AI systems.
Papers
AgentCyberRange: Benchmarking Frontier AI Systems in Realistic Cyber Ranges
Fengyu Liu, Jiarun Dai, Yihe Fan, Wuyuao Mai +10 more
The paper introduces AgentCyberRange, an open, multi-range infrastructure for measuring autonomous cyber attack capability in realistic cyber ranges, and evaluates six frontier AI systems.